Your attribution model is either helping you make better budget decisions or actively misleading you. There's no middle ground.
This decision tree cuts through the confusion. Answer a few questions about your business, and you'll know exactly which model to use, how to configure it, and which numbers to trust when platforms disagree.
Based on audits across 20+ e-commerce brands spending $3K-500K/month.
Pull your monthly conversion count from GA4. This single number determines everything downstream.
Your model: Last Click.
You don't have enough data for anything sophisticated. DDA needs 300+ conversions to function. Running it below 50 produces random attribution assignments - we've verified this across 35+ accounts.
Last click is blunt. It gives 100% credit to the final touchpoint. But at this volume, at least it's consistent and understandable.
Configuration:
What to trust:
Priority: Fix your conversion volume before worrying about attribution model sophistication. Better offer, better landing page, more traffic - any of these will get you to 50+ faster than tweaking your attribution setup.
Next step: Move to Path B when you hit 50+ conversions consistently for 3 months.
Your model: Last Click primary, DDA secondary.
DDA is starting to see patterns at this volume, but it's not reliable enough to base budget decisions on. Use both views in parallel.
Configuration:
What to trust:
Watch for: DDA giving branded search zero credit while last click gives it everything. The truth is usually between the two. If you see extreme disagreement between models, your consideration cycle is probably longer than your attribution window.
Next step: Move to Path C when you hit 300+ conversions consistently for 2 months.
Your model: Data-Driven Attribution.
You have enough data for the algorithm to identify real patterns. DDA will show you the contribution of upper-funnel channels that last click ignores.
Configuration:
What to trust:
Verification: Run a manual holdout test quarterly. Pause one channel for 2 weeks and measure true revenue impact. Compare against what DDA said the channel was contributing.
Your model: DDA with incrementality testing.
DDA is reliable at this volume. Layer incrementality studies on top for high-confidence budget allocation.
Configuration:
What to trust:
Your attribution window must match your customer's buying behavior. If it doesn't, you're cutting off conversions before they happen.
Products under $50, impulse buys, consumables, replenishment.
Products $50-200, first-time purchases, products requiring research.
Products $200+, luxury goods, high-consideration items.
Run this monthly. It takes 15 minutes. It prevents thousands in misallocated budget.
| Source | Where to Find It | What It Represents |
|---|---|---|
| Shopify revenue | Shopify Admin > Analytics > Finances | Ground truth |
| GA4 revenue | Reports > Monetization > Overview | Baseline attribution |
| Google Ads revenue | Campaigns > Conversions value column | Google's self-reported credit |
| Meta Ads revenue | Ads Manager > Purchase conversion value | Meta's self-reported credit |
| Platform total | Google + Meta + other platforms summed | Combined platform claims |
GA4 vs Shopify gap:
Platform total vs Shopify gap:
Month-over-month consistency:
Run through this checklist. Every "No" is costing you data.
| Monthly Conversions | Model | Trust Level | Reconciliation Frequency |
|---|---|---|---|
| Under 50 | Last Click | Low - directional only | Monthly |
| 50-300 | Last Click primary, DDA secondary | Medium - compare both views | Bi-weekly |
| 300-1,000 | DDA | High - verify against Shopify | Monthly |
| 1,000+ | DDA + incrementality | Highest - quarterly holdout tests | Weekly |
| Purchase Cycle | Attribution Window | Min Evaluation Period |
|---|---|---|
| Same day-3 days | 7-day click | 7 days |
| 4-14 days | 14-day click | 21 days |
| 15-30+ days | 30-day click | 30 days |
| Gap (GA4 vs Shopify) | Status | Action |
|---|---|---|
| Under 10% | Healthy | Monitor monthly |
| 10-20% | Acceptable | Investigate tracking gaps |
| Over 20% | Broken | Immediate audit |
This tree gets you to the right model with the right configuration. It tells you what to trust and what to question.
What it doesn't cover:
The Google Ads AI Agentic System ($4,997+) covers all of this. It includes the full implementation protocol, pre-built dashboard templates, server-side tracking setup guides, and the troubleshooting system built from 20+ account audits.
This decision tree gets you started. The Agentic System gets you finished.